• International Journal of Computer Science & Network Security
• /
• v.22 no.4
• /
• pp.53-58
• /
• 2022

Pneumonia is a form of acute respiratory infection that affects the lungs. According to the World Health Organization, pneumonia is the leading cause of death for children worldwide. As a result, pneumonia was the top killer of children under the age of five years old in 2015, which is 15% of all deaths worldwide. In this paper, we used CNN model architectures to compare between the result of proposed a CNN method with VGG based model architecture. The model's performance in detecting pneumonia shows that the proposed model based on VGG can classify normal and abnormal X-rays effectively and more accurately than the proposed model used in this paper.

• 한국환경농학회:학술대회논문집
• /
• 2009.07a
• /
• pp.187-199
• /
• 2009

We present real.time, rapid detection of Mycoplasma pneumonia in phosphate buffered saline (PBS) inside a Y.channel polydimethylsiloxane (PDMS) microfluidic device by means of optical fiber monitoring of latex immunoagglutination. The latex immunoagglutination assay was performed with serially diluted Mycoplasma pneumonia solutions using highly carboxylated polystyrene particles of 390nm and 500nm diameter conjugated with monoclonal anti. Mycoplasma pneumonia . Proximity optical fibers were located around the viewing cell of the device, which were used to measure the increase in 45${\b{o}}$ forward light scattering of the immunoagglutinated particles. The detection limit was less than 50 $pgml^{-1}$ both for 390nm and 500nm microspheres with the detection time less than 90 seconds.

• Journal of Veterinary Clinics
• /
• v.36 no.4
• /
• pp.185-189
• /
• 2019

Respiratory pathogens of calves including bovine parainfluenza type 3 virus (BPI3V), bovine respiratory syncytial virus (BRSV), infectious bovine rhinotracheitis virus (IBRV) and Mycoplasma spp is well-known for winter pathogens. However, there are no studies about summer pneumonia pathogens of calves in Korea. The aim of this study was to detect respiratory pathogens from calves with summer pneumonia. Eighty calves from 5 regions were chosen and their nasal swabs were used to detect respiratory pathogens with real-time PCR. Mycoplasma spp was major primary respiratory pathogens in calves with summer pneumonia. Although, the detection rates of respiratory viruses were very low, serological assays showed that respiratory viruses exist widely in farms.

• Journal of radiological science and technology
• /
• v.46 no.4
• /
• pp.277-285
• /
• 2023

In this study, pneumonia identification networks with the small number of layers were constructed by using chest X-ray images. The networks had similar trainable-parameters, and the performance of the trained models was quantitatively evaluated with the modification of the network architectures. A total of 6 networks were constructed: convolutional neural network (CNN), VGGNet, GoogleNet, residual network with identity blocks, ResNet with bottleneck blocks and ResNet with identity and bottleneck blocks. Trainable parameters for the 6 networks were set in a range of 273,921-294,817 by adjusting the output channels of convolution layers. The network training was implemented with binary cross entropy (BCE) loss function, sigmoid activation function, adaptive moment estimation (Adam) optimizer and 100 epochs. The performance of the trained models was evaluated in terms of training time, accuracy, precision, recall, specificity and F1-score. The results showed that the trained models with the small number of layers precisely detect pneumonia from chest X-ray images. In particular, the overall quantitative performance of the trained models based on the ResNets was above 0.9, and the performance levels were similar or superior to those based on the CNN, VGGNet and GoogleNet. Also, the residual blocks affected the performance of the trained models based on the ResNets. Therefore, in this study, we demonstrated that the object detection networks with the small number of layers are suitable for detecting pneumonia using chest X-ray images. And, the trained models based on the ResNets can be optimized by applying appropriate residual-blocks.

• Journal of Radiation Industry
• /
• v.18 no.3
• /
• pp.177-182
• /
• 2024

The purpose of this study is to develop a CNN based deep learning model that can effectively detect pneumonia by analyzing chest X-ray images of adults over the age of 20 and compare it with VUNO, LUNIT a commercialized AI model. The data of chest X-ray image was evaluate based on accuracy, precision, recall, F1 score, and AUC score. The CNN model recored an accuracy of 82%, precision 76%, recall 99%, F1 score 86%, and AUC score 0.7937. The VUNO model recordded an accuracy of 84%, precision 81%, recall 94%, F1 score 87%, and AUC score 0.8233. The LUNIT model recorded an accuracy of 77%, precision 72%, recall 96%, F1 score 83%, and AUC score 0.7436. As a result of the Confusion Matrix analysis, the CNN model showe FN (3), showing the highest recall rate (99%) in the diagnosis of pneumonia. The VUNO model showed excellent overall perfomance with high accuracy (84%) and AUC score (0.8233), and the LUNIT model showed high recall rate (96%) but the accuracy and precision showed relatively low results. This study will be able to provide basic data useful for the development of a pneumonia diagnosis system by comprehensively considers the perfomance of the medel is necessary to effectively discriminate between penumonia and normal groups.

• Tuberculosis and Respiratory Diseases
• /
• v.68 no.4
• /
• pp.205-211
• /
• 2010

Background: Procalcitonin is a well known marker in infection that plays a role in distinguishing between bacterial and viral infections in screening. The aim of the present study was to evaluate the role of procalcitonin in differentiating between 2009 H1N1 influenza pneumonia and community acquired pneumonia of bacterial origin, or mixed bacterial origin and 2009 H1N1 influenza infection. Methods: A retrospective observational study was performed over the 6-month winter period during the 2009 H1N1 influenza pandemic. Ninety-six patient-subjects were enrolled, all of whom had been diagnosed with community acquired pneumonia in emergency department during the study period. On admission, laboratory studies were performed, which included 2009 H1N1 influenza real-time polymerase chain reaction of nasal secretions and procalcitonin on serum; the laboratory values were compared between the study groups. Receiver operating characteristic curve analyses were performed on the resulting data. Results: Compared to those with bacterial or mixed infections (n=62) and bacterial pneumonia with confirmed organisms (n=30), patients with 2009 H1N1 pneumonia (n=34) were significantly more likely to have low procalcitonin levels (p=0.008, 0.001). Using cutoff of value >0.3 ng/mL, the sensitivity and specificity of procalcitonin for detection of patients with confirmed bacterial pneumonia were 76.2% and 60.6%, respectively. A significant difference in procalcitonin was found between 2009 H1N1 pneumonia and pneumonia caused by mixed influenza viral and bacterial infections (0.15 [0.05~0.84] vs. 10.3 [0.05~22.87] ng/mL, p=0.045). Conclusion: Serum procalcitonin measurement may assist in the discrimination between pneumonia of bacterial and of 2009 H1N1 influenza origin. High values of procalcitonin suggest that bacterial infection or mixed infection of bacteria and 2009 H1N1 influenza is more likely.

• Korean Journal of Radiology
• /
• v.21 no.10
• /
• pp.1150-1160
• /
• 2020

Objective: To describe the experience of implementing a deep learning-based computer-aided detection (CAD) system for the interpretation of chest X-ray radiographs (CXR) of suspected coronavirus disease (COVID-19) patients and investigate the diagnostic performance of CXR interpretation with CAD assistance. Materials and Methods: In this single-center retrospective study, initial CXR of patients with suspected or confirmed COVID-19 were investigated. A commercialized deep learning-based CAD system that can identify various abnormalities on CXR was implemented for the interpretation of CXR in daily practice. The diagnostic performance of radiologists with CAD assistance were evaluated based on two different reference standards: 1) real-time reverse transcriptase-polymerase chain reaction (rRT-PCR) results for COVID-19 and 2) pulmonary abnormality suggesting pneumonia on chest CT. The turnaround times (TATs) of radiology reports for CXR and rRT-PCR results were also evaluated. Results: Among 332 patients (male:female, 173:159; mean age, 57 years) with available rRT-PCR results, 16 patients (4.8%) were diagnosed with COVID-19. Using CXR, radiologists with CAD assistance identified rRT-PCR positive COVID-19 patients with sensitivity and specificity of 68.8% and 66.7%, respectively. Among 119 patients (male:female, 75:44; mean age, 69 years) with available chest CTs, radiologists assisted by CAD reported pneumonia on CXR with a sensitivity of 81.5% and a specificity of 72.3%. The TATs of CXR reports were significantly shorter than those of rRT-PCR results (median 51 vs. 507 minutes; p < 0.001). Conclusion: Radiologists with CAD assistance could identify patients with rRT-PCR-positive COVID-19 or pneumonia on CXR with a reasonably acceptable performance. In patients suspected with COVID-19, CXR had much faster TATs than rRT-PCRs.

• Tuberculosis and Respiratory Diseases
• /
• v.80 no.4
• /
• pp.358-367
• /
• 2017

Background: Bacterial pneumonia occurring after respiratory viral infection is common. However, the predominant bacterial species causing pneumonia secondary to respiratory viral infections other than influenza remain unknown. The purpose of this study was to know whether the pathogens causing post-viral bacterial pneumonia vary according to the type of respiratory virus. Methods: Study subjects were 5,298 patients, who underwent multiplex real-time polymerase chain reaction for simultaneous detection of respiratory viruses, among who visited the emergency department or outpatient clinic with respiratory symptoms at Ulsan University Hospital between April 2013 and March 2016. The patients' medical records were retrospectively reviewed. Results: A total of 251 clinically significant bacteria were identified in 233 patients with post-viral bacterial pneumonia. Mycoplasma pneumoniae was the most frequent bacterium in patients aged <16 years, regardless of the preceding virus type (p=0.630). In patients aged ${\geq}16years$, the isolated bacteria varied according to the preceding virus type. The major results were as follows (p<0.001): pneumonia in patients with influenza virus (type A/B), rhinovirus, and human metapneumovirus infections was caused by similar bacteria, and the findings indicated that Staphylococcus aureus pneumonia was very common in these patients. In contrast, coronavirus, parainfluenza virus, and respiratory syncytial virus infections were associated with pneumonia caused by gram-negative bacteria. Conclusion: The pathogens causing post-viral bacterial pneumonia vary according to the type of preceding respiratory virus. This information could help in selecting empirical antibiotics in patients with post-viral pneumonia.

• Korean Journal of Veterinary Service
• /
• v.38 no.1
• /
• pp.31-36
• /
• 2015

Two cats were obtained from a cat kennel. Over the previous 7 days, the cats had shown cough, anorexia, depression and nasal discharge. In this study, the consensus PCR was able to detect successfully Mycoplasma species in nasal swab samples of the cats. To identify feline mycoplasma species from the lung tissue of the cats with pneumonia, Mycoplasma species-specific PCR reactions were conducted. As the results, we could identify M. felis by the positive amplified DNAs. On the other hand, we could not detect any positive reactions with the PCR reaction for M. arginini, M. canis, M. edwardii, M. cynos, M. gateae, M. maculosum, M. molared, M. opalescens, M. spumans and Mycoplasma HRC-689. In conclusion, we detected M. felis from the kenneled cats with pneumonia. We suggested that this consensus PCR would be useful and effective for monitoring Mycoplasma species in various kinds of animals including cats. The application of preceding consensus PCR before the species-specific PCRs may be the most recommended strategy for the identification of Mycoplasma spp.

• Clinical and Experimental Pediatrics
• /
• v.52 no.3
• /
• pp.315-323
• /
• 2009

Purpose : This study aimed to perform a systematic review of the reports on Mycoplasma pneumoniae pneumonia in the last 30 years (1980-2006) to investigate the intervals between outbreaks, change in the peak incidence age, and diagnostic methods. We also aimed to validate the proper diagnostic criteria for M. pneumoniae pneumonia. Methods : We reviewed 62 original articles on M. pneumoniae pneumonia in Korean children. We analyzed the annual or seasonal variation, study areas, patient age, journal names, and the date of each report. Further, we checked the methods and criteria used for the diagnosis of M. pneumoniae pneumonia. We also confirmed the proper mycoplasma antibody cutoff using the mycoplasma IgM titer as the gold standard. Results : In the last 30 years, epidemic outbreaks of M. pneumoniae pneumonia occurred every 3 years, except in 1993-1994 and 1996-1997. Seasonal variations were also present and were most prevalent in October and November. The number of preschool children, especially those aged 3 years or younger, with M. pneumoniae pneumonia has increased (P<0.05). The mycoplasma antibody titer of 1:640 or greater was appropriate for diagnosing M. pneumoniae pneumonia, with an acceptable sensitivity and specificity of detection. Conclusion : We analyzed the results of studies on M. pneumoniae pneumonia in Korean children during the last 30 years. Infection in younger children is increasing, and further research is required to reveal the major cause of the changing epidemics.